Body oncologic imaging subspecialty training a curriculum based on the experience in a tertiary cancer center

Abstract

Purpose:

To describe the structure of a dedicated body oncologic imaging fellowship program. To summarize the numbers and types of cross-sectional imaging examinations reported by fellows.

Methods:

The curriculum, training methods, and assessment measures utilized in the program were reviewed and described. An educational retrospective analysis was conducted. Data on the number of examinations interpreted by fellows, breakdown of modalities, and examinations by disease management team (DMT) were collected.

Results:

A total of 38 fellows completed the fellowship program during the study period. The median number of examinations reported per fellow was 2296 [interquartile range: 2148 – 2534], encompassing all oncology-relevant imaging modalities: CT 721 [646–786], MRI 1158 [1016–1309], ultrasound 256 [209–320] and PET/CT 176 [130–202]. The breakdown of examinations by DMT revealed variations in imaging patterns, with MRIs most frequently interpreted for genitourinary, musculoskeletal, and hepatobiliary cancers, and CTs most commonly for general staging or assessment of nonspecific symptoms.

Conclusion:

This descriptive analysis may serve as a foundation for the development of similar fellowship programs and the advancement of body oncologic imaging. The volume and diversity of examinations reported by fellows highlights the comprehensive nature of body oncologic imaging.

1. Introduction

Imaging plays a crucial role in the screening, diagnosis, treatment, and follow-up of cancer patients. Advances in imaging technology have enabled increasingly accurate tumor characterization, treatment response monitoring, and recurrence detection. These same advances have also resulted in increasing complexity of cancer imaging, requiring dedicated expertise to ensure accurate [1–4] and timely [5] interpretation of imaging studies.

In recognition of this need, a few centers offer dedicated fellowship programs for radiologists seeking to gain expertise in body oncologic imaging. These programs expose radiologists to a diverse range of oncologic scenarios and cover the commonly employed modalities in cancer patients including CT, MRI, PET/CT and ultrasound. Body oncologic imaging programs also typically transcend the domains of other common radiology subspecialties including abdominal, thoracic, musculoskeletal imaging and nuclear medicine. This mirrors the complexity of cancer as a multisystem disease, and the clinical reality that patients with cancer are increasingly managed in a multidisciplinary environment.

Radiology training programs are available in more than 200 institutions in the U.S. Only seven institutions offer a clinically oriented “cancer imaging fellowship” of at least 1 year duration in 2022. Merely two of them were added in the last 5 years [6].

The call for more oncologic imaging-specific training, including a structured and standardized curriculum, has previously been recognized [6–10]. Structure and standardization are necessary to ensure trainees receive a comprehensive understanding of disease and treatment characteristics in various malignancies. As the need for cancer imaging specialists increases, so does the need to equip radiology trainees with a fundamental knowledge base upon which to build throughout their career. A solid foundation enables them to adapt with future advances in both oncologic diagnosis and therapeutic interventions.

In this paper, we report the structure and content of a dedicated oncologic imaging fellowship program. This report is based on the experience over the last 5 years. We summarize the curriculum and training methods utilized, and the assessments and evaluations employed. Qualifications and characteristics of these fellows are also described in detail.

2. Methods

As an educational retrospective study not involving protected health information (PHI), this analysis was exempt from institutional review board review.

2.1. Data extraction and analysis

Imaging examination counts for different modalities and examination codes were extracted from the institutional radiological information system from the fellow classes 2017–2021 (July 1, 2017 - June 30, 2022). Data from two fellows were excluded due to incomplete fellowship duration. Data were summarized as median and interquartile range. Due to the descriptive nature of this paper, only summary statistics were performed.

2.2. Fellowship details

2.2.1. Fellow selection

The fellowship program is open to radiologists who have completed a diagnostic radiology residency program in the U.S. or abroad and are board-certified or board-eligible by the American Board of Radiology. Fellows are selected based on their academic achievements, professional accomplishments, and potential for future contributions to the field of oncologic imaging. In addition, fellows must demonstrate a strong interest in oncologic imaging and commit to a 12-month, full-time training program.

2.2.2. Curriculum.

The curriculum is designed to provide fellows with a comprehensive and in-depth understanding of oncologic imaging. Fellows gain exposure to a diverse range of cancers, including both solid and hematologic malignancies in pediatric and adult patients. Fellows also receive training in a range of imaging modalities.

The core rotations are divided into MRI (26 weeks), CT (12 weeks), ultrasound (4 weeks), and PET/CT (4 weeks). The emphasis on MRI was based on two main factors: (1) Feedback from earlier fellowship classes highlighted a limited exposure to oncologic MRI in many residency programs, and (2) MRI is increasingly used in oncologic imaging. This increase is driven in part by the multiple Reporting and Data Systems (-RADS) utilizing MRI that are now used in standard of care for several cancers (e.g. PI-RADS, O-RADS, LI-RADS, etc.).

In addition to the core rotations, fellows have the opportunity to participate in elective clinical rotations. Popular clinical electives are breast imaging, neuro-oncologic imaging, musculoskeletal imaging and interventional radiology. These rotations are offered both in-house or at a partner hospital (e.g. tertiary musculoskeletal hospital). In combination or as an alternative, the elective time may be used for scientific research.

The program does not include unsupervised or on-call work. Included are some evening shifts on weekdays and 12-hour day shifts on weekends. Fellows, under faculty supervision, are responsible for timely reporting of inpatient and oncologic emergency room ultrasound and CT examinations. This experience provides exposure to imaging of oncologic emergencies and postoperative complications. Fellows also occasionally interpret conventional radiographs during these shifts.

Academic departmental conferences are held on a weekly basis where fellows present interesting cases, journal articles or give didactic lectures. Fellows also are encouraged to participate in research projects with mentorship from subspecialized oncologic radiologists. These academic activities allow them to acquire in-depth knowledge about the latest oncologic imaging topics and foundational scientific methodology.

2.2.3. Training methods.

Fellows receive hands-on training in all aspects of oncologic imaging, including image acquisition, interpretation, and reporting. They first review each exam independently and write a preliminary report, followed by a side-by-side review and discussion of relevant imaging features and related literature with a designated faculty member. The faculty member then makes the necessary edits to the preliminary findings and issues a final co-signed report, bearing the names of both fellow and faculty member.

Fellows regularly present interesting cases at departmental didactic conferences. They also participate in multidisciplinary conferences and rounds at least once a week. In this multidisciplinary setting they present cases and collaborate to develop treatment plans together with attending oncologists, surgeons, pathologists, and other specialists. The goal of this approach is to provide fellows with a comprehensive understanding of the role of imaging in cancer diagnosis, treatment planning, and follow-up care.

The number of cases varies depending on the format of the conference. Some conferences are limited to one hour, while others are open-ended, lasting typically anywhere between one to three hours. In general, fellows will present 2–3 cases in the beginning of their fellowship, with the aim that they are able to present the full list of cases for a given conference by the end of the fellowship. Following the same principle of individualized training and progression of responsibilities, fellows may choose to participate as an observer for their first one or two conferences, whereas they are expected to be able to lead the case presentation and actively participate in the multidisciplinary discussion later in their fellowship.

2.2.4. Examination Assignments and Numbers.

Multidisciplinary patient care is provided via disease management teams (DMT). These may be systems-based, such as genitourinary or gynecologic, or disease-based, such as the melanoma or lymphoma DMT. Examinations of the “general” category are tailored to provide a general oncologic assessment. This category is assigned irrespective of the DMT. Examples are characterization of liver lesions, assessment of treatment complications or investigation of nonspecific symptoms in cancer survivors.

Fellows receive broad guidance on the number of cases to review per day for each rotation. The number of examinations assigned to each fellow varies depending on several factors, including the complexity of the examination, the fellow’s level of experience, and the number of patients being imaged.

Daily examinations are assigned according to the current modality-based rotation of the fellow, stratified by DMT. Stratified means they receive one modality with assignments aiming for an approximately equal share of DMTs available on the given worklist [11]. Numbers are adjusted over the year to reflect increasing skill and experience.

2.2.5. Assessment and Evaluation

Fellows are formally evaluated on a regular basis throughout the fellowship program by all faculty using electronic questionnaires. The questionnaires aim to capture clinical knowledge, quality of reports, professional conduct and progression during fellowship. The formal evaluation process ensures that they are meeting the program’s goals and objectives. Individual evaluations are based on the fellows’ performance in clinical settings, didactic sessions, and research projects. Fellows also receive direct feedback from their mentors and faculty in addition to two formal evaluations by the fellowship program directors at approximately 3 months and 8 months into the fellowship program.

3. Results

Over the 5-year period, 38 fellows completed the oncologic imaging fellowship program. Fig. 1 shows the distribution of the total number of examinations interpreted per fellow each year: The median number of examinations per fellow was 2330 in 2017, 2268 in 2018, 2185 in 2019 (slightly lower due to the pandemic months), 2446 in 2020, and 2344 in 2021.

Fig. 1.

Boxplot graph of total examination count per fellow per year. The middle line represents the median, upper and lower hinges correspond to the first and third quartile, respectively. The whiskers extend to +/− 1.5 x quartile from the hinges (Tukey-method).

Table 1 shows the breakdown of imaging modalities interpreted by the fellows during their training. In accordance to the duration of the core rotations, the majority of examinations were MRIs, with a median of 1158 per fellow. This was followed by CTs, with a median of 721 per fellow, ultrasound with 256 and PET-CTs with a median 176 per fellow.

Table 1.

Summary of interpreted examinations grouped by modality.

Modality	Median	IQR
MR	1158	1016–1309
CT	721.0	646–786
US	256	209–320
PET-CT	176	130–202

Table 2 presents the number of examinations interpreted by each fellow according to disease management team (DMT). The general category includes examinations performed for general oncologic assessment, regardless of DMT. The most commonly interpreted MRI examinations were from the genitourinary (GU), musculoskeletal/sarcoma, hepatopancreaticobiliary (HPB), gynecologic (GYN) and gastro-intestinal (GI) DMTs, followed general examinations, and pediatric, chest & vascular, hematologic as well as the melanoma and thyroid DMTs. The most common CTs were general examinations, followed by HPB, GU, chest & vascular, musculoskeletal/sarcoma, GYN, GI, hematologic, melanoma, pediatric, and thyroid DMT.

Table 2.

Summary of interpreted examinations grouped by disease management team (DMT).

DMT	MR: median	MR: IQR	CT: median	CT: IQR	PT: median	PT: IQR
GU	377	345–384	83	81–86	15	13–21
MSK/Sarcomas	243	223–250	47	43–63	6	5–7
HPB	196	180–201	105	88–107	10	8–13
GYN	139	122–146	40	38–47	8	8–9
GI	73	66–121	27	26–47	10	7–10
General	71	70–73	326	267–373	40	38–48
Pediatric	15	10–19	5	4–7	4	3–4
Chest & cardiovascular	10	9–16	71	57–77	32	24–33
Hematologic malign.	9	7–17	16	14–36	28	26–44
Melanoma	9	7–9	7	4–8	5	4–7
Thyroid	4	3–5	5	4–6	15	11–17

The variation in examination volume across different DMTs reflects the diverse nature of oncologic imaging. Each disease type requires a unique imaging approach and a different set of imaging tools. The differences in examination frequency across DMTs also reflect the prevalence of different types of cancer, as well as the distribution of cancer types across different populations. A more extensive discussion of the relationship between DMTs and imaging patterns follows in the next section.

Table 3 provides a comprehensive list of the most commonly interpreted imaging protocols. The table encompasses CT and MRI protocols utilized for different disease types and clinical questions. The exhaustive list highlights the broad range of imaging techniques required for comprehensive oncologic imaging and emphasizes the depth of expertise gained by fellows during their training. The breadth of protocols also underscores the importance of subspecialty training in oncologic imaging. A detailed discussion of various image protocols and their clinical applications follows below.

Table 3.

Summary of commonly interpreted imaging protocols. A forward slash indicates an “or”, the ampersand sign “&” is used for “and”.

Exam type	Median	IQR
MR Prostate	267	224–271
CT CAP	251	238–254
MR Abd. & Pel.	243	208–244
CT Abd. & Pel.	205	194–219
PET-CT	176	130–202
MR Joint/Extremity	174	165–182
CT Chest (incl. PE)	173	158–176
MR Liver	114	112–144
MRCP	90	88–101
MR GYN pelvis	78	72–83
CT Liver/Kidney/Urogram	50	47–52
MR Kidney/Urogram	49	44–49
MR Rectum	46	44–94
MR Pelvis general	36	33–38
MR Whole body	34	27–37
MR Chest	25	18–30
CT Angiogram	19	18–21
CT Joint/Extremity	13	13–18
MR Abdomen	10	9–11
CT Pelvis	8	7–8

4. Discussion

This study summarizes the hands-on reporting experience of sub-specialty trainees within a dedicated body oncologic imaging program. As imaging plays a crucial role in the diagnosis, staging, and management of cancer, specialized knowledge and expertise in this area are critical to ensure accurate interpretation and optimal patient outcomes. Indeed, previous studies have shown that subspecialty-trained radiologists demonstrate superior diagnostic accuracy and efficiency in oncologic imaging [1–4,12]. The present study underscores the importance of such training, emphasizing the imaging needs of patients with cancer which may deviate from those with other diseases.

The anatomic coverage often differs from other more localized conditions and transcends conventional radiology department sections. For example, oncologic CT exams, not only in routine clinical practice but also in clinical trials, frequently cover the chest, abdomen and pelvis (the 2nd most interpreted examination type overall in our study). A combined review and single report (as opposed to two separate reports for chest and abdominopelvic portions) is advantageous for comprehensive assessment of disease spread, for purposes of staging and treatment response assessment. Whole-body imaging with PET and more recently MRI is also commonly applied in the oncologic setting.

The example of PET and MRI also highlight the importance of a multimodality approach. An oncologic radiologist trained and proficient in multiple modalities can use a PET/MRI for both lesion characterization and whole-body staging. At our institution, contemporaneous examinations of different modalities from the same patient are assigned to the same radiologist for integrated interpretation and reporting, e.g., MRI for local extent of disease evaluation and CT of chest, abdomen and pelvis for systemic staging [5].

These needs and trends are reflected in the experience obtained in the fellowship program, where fellows were exposed to a multimodality approach (predominantly MRI, CT, and PET) covering multiple sub-specialty- or anatomy-based radiology sections (e.g., chest, body, musculoskeletal and pediatrics). Of note, the fellowship does not include a dedicated rotation for conventional radiography, since radiology trainees usually encounter a sufficient volume of radiographs during their residency training. Radiographs are also not considered primary modalities for assessment of cancer presence or distribution. When relevant, for example for bone tumor characterization, radiographs are reviewed concurrently with cross-sectional imaging.

The proportion of exams from different DMTs reflects the diverse utilization rates across the spectrum of oncologic imaging. The reasons for utilization patterns are complex and depend on a multitude of factors such as disease prevalence, patient demographics, anatomical location, availability of modalities and guideline recommendations. Each disease type requires a unique imaging approach and a different set of imaging tools. For example, the predominance of MRI examinations in the GU DMT is in part due to a convergence of two factors: Prostate cancer has a very high prevalence. In addition, changes to major guidelines in the past decade [13] made MRI an important pillar in the management of patients with suspected or established prostate cancer. Similarly, the predominance of CTs in the Chest DMT reflects the fact that lung tumors are generally better visualized with CT. Exceptions are for example marginal cases where MRI may be needed to determine chest wall invasion [14].

As part of the fellowship experience, fellows routinely encounter examinations from patients enrolled in clinical trials. Contact with trial examinations provides a unique opportunity to directly study the effects and complications of novel surgical, radiotherapy and systemic treatments during their early stages of development [15]. By participating in this research, fellows gain expertise in emerging fields and contribute to the investigation of breakthrough therapies.

The present study has several limitations that need to be acknowledged. The study is retrospective in nature, and as such is subject to potential biases and limitations inherent in this type of analysis. The observational retrospective design precluded an assessment of improvement in performance over the 12-month-period. The study was conducted at a single institution, and may thus not be representative of imaging patterns and practices in other settings, whose patient populations and DMT-divisions may differ. Moreover, the size of the respective clinical groups may be larger or smaller, resulting in different referral patterns. For example, the high expertise in treating prostate cancer at our institution results in a large number of prostate MRIs. The summary of volume by protocols is limited as some protocols may fall under different DMT’s. This limitation has been mitigated by also including a breakdown of volume by DMT.

Financial and access-to-care constraints have not been considered, which may impact the case-mix and required skills, i.e. training needs, for radiologists in different settings. However, fellowship training itself is an important pillar to improve access to subspecialized care [10] by means of graduates practicing in the community.

Only data from the dedicated diagnostic body imaging fellowship were analyzed. The institution offers additional fellowships, including a combined breast and body imaging fellowship, molecular imaging and therapy fellowship, and interventional fellowship, which were not included in the current analysis.

The experiences and data of this dedicated fellowship program, we hope, may spark further publications on oncologic imaging training. These data may serve as guidance for the development and refinement of similar programs. By increasing oncologic imaging fellowships, our field of radiology can make an important contribution to the improvement of patient care.